Dynatrace Synthetic Monitoring Pricing and Cost per Action

Dynatrace charges $1.00 per 1,000 synthetic HTTP requests and $4.50 per 1,000 synthetic browser actions, which works out at $0.001 per request and $0.0045 per action. That browser rate is the second lowest headline number in the synthetic monitoring market, and it is also the most expensive real rate you can pay for any journey longer than two steps, because Dynatrace is the only major vendor that bills a browser check per step rather than per run. A three step clickpath costs $0.0135 a run against Datadog's flat $0.012, and a ten step checkout costs $0.045.

Every other synthetic vendor sells you a run. Datadog, Checkly, Grafana Cloud, New Relic and AWS all charge one price whether your browser script loads a homepage or walks a fifteen step purchase flow. Dynatrace charges for what the script does. That single design decision flips the vendor from cheapest to dearest somewhere between the second and third step of your journey, and nothing on the pricing page tells you where the line is. This post works out where it is.

The rate card, verified

These figures were read on Dynatrace's own pricing page in August 2026. They are list prices under the Dynatrace Platform Subscription model, and the vendor's documentation notes your negotiated rate card may differ.

Two of those lines cross-check each other, which is a useful sanity test when you are reading any vendor's pricing page. Infrastructure Monitoring is quoted both as $29 a month and as $0.04 an hour. Multiply $0.04 by 730 hours and you get $29.20, so Dynatrace is quoting a 730 hour month and rounding down. Full-Stack at $0.01 per memory-GiB-hour across 8 GiB for 730 hours is $58.40, quoted as $58. The arithmetic holds, which tells you the units are being read correctly.

Why per action changes everything

Dynatrace's documentation is unusually specific about what an action is, and the rules are worth knowing because they are also the cost control.

A step that triggers one or more web requests counts as one synthetic action. A step that fires three XHR requests is still billed as a single action, so chatty pages do not cost extra. A step that triggers no web request at all counts as zero: scrolls, keystrokes and clicks that do not hit the network are free. XHR or fetch calls the browser makes automatically after a page load are also free unless a user action explicitly triggered them. An action is billed the moment the step executes, whether it passes or fails, but steps after a failed action never execute and so never bill.

That last set of rules is the lever. A twenty step journey in which only six steps actually talk to a server costs the same as a six step journey. Filling in a form field by field is free until you submit it. At a five minute interval from one location, a twenty step journey with six billable steps runs $233.28 a month, against $777.60 if every step hit the network. The difference is $544 a month on one monitor, decided entirely by how the script is written.

What a browser monitor actually costs per month

A 30 day month holds 43,200 minutes, which is the number that converts any per unit rate into an invoice. Runs multiply by billable steps, then by locations. One browser monitor, one location:

Dynatrace's own documentation carries a worked example that lands in the middle of that table: three billable steps, executed four times an hour across three locations over a 720 hour month, is 25,920 synthetic actions, or $116.64. Our arithmetic reproduces it exactly, which is a good sign that the model above is the model they bill.

The crossover point nobody publishes

Here is the comparison that decides whether Dynatrace is cheap or expensive for you. All the rivals below charge per run, so their number does not move with journey length. Dynatrace's does.

So Dynatrace wins on price only for single step browser checks, which is the one case where you probably did not need a browser at all. For anything that looks like a real user journey, meaning a login, a search, an add to cart and a checkout, it is the most expensive option on this list. At a five minute interval from one location, a ten step journey is $388.80 a month at Dynatrace against $103.68 at Datadog, $54.00 at Checkly, $43.20 at Grafana and $10.37 at AWS.

None of that makes the model wrong. Charging for work performed is arguably the fairest way to meter a browser check, and it is the only pricing on the market that rewards you for writing an efficient script. It just means the headline rate is not comparable to anyone else's headline rate, and quoting it next to Datadog's $0.012 without the step multiplier is how monitoring budgets get set at a third of their eventual size.

HTTP monitors bill per request, not per monitor

The same logic applies on the API side, and it is missed more often because the unit sounds like a run. Dynatrace bills synthetic requests, so a multi request HTTP monitor bills each request in the sequence. A monitor that authenticates, then calls an endpoint, then verifies a follow up resource is three requests, not one.

Ten API monitors checked every five minutes from three locations is 259,200 requests, or $259.20 a month if each monitor makes one request. Make it three requests each and the same setup is $777.60. That is a three fold move driven by a detail most people would describe as "the monitor checks a couple of endpoints".

How to keep a Dynatrace synthetic bill down

Four things move the number, in rough order of leverage.

Count your billable steps, not your steps. Open a clickpath and mark which steps produce a web request. Everything else is free. Combining two navigation steps into one, or replacing a scripted form fill with a direct POST, removes actions permanently.

Pick the interval against your error budget, not by habit. Worst case detection delay is one full check interval. Against a 99.9% monthly target, which allows 43 minutes 12 seconds of downtime, a five minute interval spends about 12% of the whole month's budget before anyone is paged. At 99.99%, where the budget is 4 minutes 19 seconds, a five minute interval has already spent 116% of it, so the target is not one you can honestly report against at that frequency. Our error budget calculator does that arithmetic both ways.

Split the deep journey from the availability check. Most teams run one expensive clickpath every minute when what they actually want is a cheap availability check every minute and the full journey every fifteen. A twelve step journey at a fifteen minute interval is $155.52 a month. The same journey every minute is $2,332.80.

Do not add locations without a reason. Every location multiplies the whole bill. Three is usually enough to distinguish a real outage from a regional network problem, and Dynatrace, unlike AWS, charges the same rate wherever the check runs from.

Is Dynatrace synthetic monitoring worth it?

If you already run Dynatrace for APM, adding synthetics keeps everything on one rate card and one trace, and the per action model is defensible. If you are buying synthetic monitoring on its own, the pricing is hard to justify: you are paying platform rates for check execution, and the platform is the part you are not using.

The wider point is about which meter a vendor picks. Dynatrace meters browser work per action, Datadog and Checkly per run, Grafana per probe-minute of run time rounded up, Better Stack per Playwright minute, and AWS per canary run at a rate that varies by region. Five vendors, five units, and the unit decides which configuration is expensive. Our synthetic monitoring pricing comparison holds all of them in one table with a calculator, and the Datadog Synthetics pricing breakdown covers the per run model in the same detail as this post covers per action.

There is a case for stepping outside metered pricing entirely. If what you need is to know quickly that a site, an API, a certificate or a scheduled job has stopped working, a flat per monitor plan removes the trade between cost and detection speed, because the interval stops being a billing decision. That is the model we run: every plan includes 30 second checks from six regions with no per run charge. If you need scripted multi step journeys, we do not do those, and the Checkly alternative page is a more useful comparison than this one.

One last practical note. If you are running these checks partly to evidence an availability commitment to customers, the monitoring record is only one input; the availability criteria sit alongside a couple of dozen other requirements, and there are tools that map your controls to SOC 2 and ISO 27001 so the uptime evidence you are already collecting gets filed against the right one. Worth knowing before an auditor asks for it rather than during.

Common questions

How much does Dynatrace synthetic monitoring cost? HTTP monitors are $1.00 per 1,000 synthetic requests and browser monitors are $4.50 per 1,000 synthetic actions, at list price. A three step browser journey checked every five minutes from one location costs $116.64 a month. A single request HTTP monitor at the same interval costs $8.64.

What is a synthetic action in Dynatrace? A synthetic action is one step of a browser monitor or clickpath that triggers at least one web request. Steps that fire several requests still count as one action, and steps that trigger no request, such as scrolling or typing, count as zero. Actions bill on execution whether the step passes or fails.

Is Dynatrace cheaper than Datadog for synthetic monitoring? Only for very short browser journeys. Dynatrace's $0.0045 per action beats Datadog's $0.012 per run at one or two billable steps, then loses from three steps onward. For HTTP checks Dynatrace is $0.001 per request against Datadog's $0.0005 per run, so Datadog is half the price on single request checks.

Does Dynatrace charge for failed synthetic checks? Yes. An action bills as soon as the step executes, regardless of the result. The only relief is that steps queued after a failed action never run, so a journey that breaks at step two does not bill steps three through ten.